Flight simulators are used by commercial airlines and air forces to train their pilots to face various types of situations. A simulator is capable of simulating various functionalities of an aircraft, and of reproducing various operational conditions of a flight (e.g. takeoff, landing, hovering, etc.). A trainee (e.g. a pilot performing a training session) interacts with the simulator to control various functionalities of the simulated aircraft during a simulation executed by the simulator. Similarly, an instructor (e.g. an experienced pilot) may interact with the simulator for various purposes, including controlling a simulation currently executed by the simulator, creating or updating simulation scenarios, controlling the simulation environment of a trainee, etc.
A particular simulation scenario executed by the simulator generally consists of a set of pre-determined events occurring during the execution of the particular simulation scenario by the simulator. The results of the simulation scenario are exported as a log file comprising simulation data and corresponding times of occurrence for the simulation data. The evaluation of a trainee who has performed the simulation scenario is based on an interpretation of the log file by an instructor, and on visual subjective observations performed by the instructor during the simulation.
One issue related to the evaluation is its subjectivity, since two different instructors may make different visual observations and interpret the same log file differently. Another issue related to the simulation scenario is its lack of flexibility, since a particular simulation scenario may only be varied by having the instructor manually introduce new challenges for the trainee in the course of the simulation. Furthermore, it is not possible to evaluate objectively a trainee's progress and difficulties, since the evaluation is only based on the log file currently at hand, for which an interpretation by an instructor is completely subjective.
One solution to this issue is to define events having at least one rule which can be measured or verified, and to further define a target value of the rule. For each event, the values of the corresponding rules are measured during the execution of a simulation according to a lesson plan comprising the events, and the measured values are compared to their target values. Furthermore, the definition of events and corresponding rules may be automated, using simulation data collected from previously executed simulations.
Therefore, there is a need for a simulation server capable of creating events of a lesson plan based on simulation data statistics.